#3 closed defect (fixed)
CholeskiInverse doesn't return the inverse matrix of the argument
Reported by: | César Pérez Álvarez | Owned by: | danirus |
---|---|---|---|
Priority: | highest | Milestone: | |
Component: | Math | Version: | head |
Severity: | major | Keywords: | TOL |
Cc: |
Description
You can execute this example to see that CholeskiInverse doesn't return the
inverse matrix of the argument.
Matrix A = SetMat(SetOfSet
(
SetOfReal(1/4,1/8,1/9),
SetOfReal(1/8,1/8,1/9),
SetOfReal(1/9,1/9,1/9)
));
See that A is positive defined
Matrix A_ = CholeskiInverse(A);
You can see that this product doesn't return the identity matrix of dim 3
Matrix U = A*A_;
Change History (4)
comment:1 Changed 21 years ago by
Status: | new → assigned |
---|
comment:2 Changed 21 years ago by
Resolution: | → fixed |
---|---|
Status: | assigned → closed |
comment:3 Changed 21 years ago by
Component: | Matrix → TYPE_Matrix |
---|---|
Keywords: | Functions removed |
product: | TOL Functions → TOL |
Moved to TOL product.
comment:4 Changed 18 years ago by
bug_file_loc: | → http://cvs.tol-project.org/viewcvs.cgi/tol_tests/tol/Bugzilla/bug_000003 |
---|
Note: See
TracTickets for help on using
tickets.
CholeskiInverse behavior, in matgra.cpp, has been rewritten to avoid confusing
internal casts in Win32 platform.
I have seen some complex and confused behavior in classes hierarchy of
BMatrix<Any>. I think that all of them should be rewritten.